This study examined the relationship between religiosity in 175 psychiatric inpatients as measured by the subscales of the Duke University Religion Index (DUREL) and sociodemographic (age, sex, and race), clinical (primary diagnosis, suicidality, and psychotic symptoms), and outcome (length of stay [LOS] and readmission rates) measures. Psychosis was assessed by Clinician-Rated Dimensions of Psychosis Symptom Severity (CRDPSS) scale. Bivariate and multivariate analyses were used to examine the association between the DUREL subscales and the outcome measures. High scorers on the nonorganized religiosity subscale were less likely to have psychosis (47% vs. 52%; p < 0.05) but had greater psychosis severity (mean ± SD, 14.5 ± 5 vs.12.4 ± 6; p < 0.05), as measured by the CRDPSS scale, and significantly longer LOS (mean ± SD, 8.3 ± 3.8 vs. 6.9 ± 3.4; p < 0.05). Conversely, they were less likely to report previous suicide attempts than low scorers (p < 0.05). These results suggest that a brief measure of religious activities may identify psychiatric inpatients at greater risk for psychosis, suicidality, and longer hospitalizations.